I am a book publisher & I love technology. It can empower people. I have been using LLM chatbots since they became widely available. I regularly test machine translation at our publishing house in collaboration with our translators. I have just completed two courses in artificial intelligence and machine learning at my alma mater, Masaryk University, and I am training my own experimental models (for predicting bestsellers :). I consider machine learning to be a remarkable invention and catalyst for progress. Despite all this, I have my doubts.
I’m with you—-I think you did a good job of summarizing all the places that LLMs are super practical/useful, but agreed that for prose (as someone who considers themselves a proficient writer), it just never seems to contribute anything useful. And those who are not proficient writers, I’m sure it can be helpful, but it certainly doesn’t contribute any new ideas if you’re not providing them.
For things like coding LLMs are useful and DEVONThink's recent AI integrations allow me to use local models as something like an encyclopedia or thesaurus to summarize unfamiliar blocks of text. At best I use it like scratch paper.
I formed the habit of exporting entire chats to Markdown and found them useless. Whatever I found of useful from a given response either sparked a superseding thought of my own or was just a reiteration of my own intuitive thoughts.
I've moved from ChatGPT to Claude. The results are practically the same as far as I can tell (although my gut tells me I get better code from Claude) but the I think Anthropic have a better feel for response readability. Sometimes processing a ChatGPT response is like reading a white paper.
Other than that, LLMs get predictable to me after a while and I get why people suspect that they're starting to plateau.
I avoided cell phones too when they first came out. I didn't want the distraction or "digital leash". Now it's a stable fixture in my life. Some technology is simply transformational and is just a matter of time until almost everyone comes to accept it at some level. Time will tell if AI breaks through the hype curve but my gut feeling is it will within 5 years.
You know that teammate that makes more work for everyone else on the team because they do what they are asked to do but in the most buggy and incomprehensible way, that when you finally get them to move on to another team and you realize how much time you spent corralling them and fixing their subtle bugs and now when they are gone work doesn't seem like so much of a chore.
My writing style is pretty labor intensive [0]. I go through a lot of drafts and read things out loud to make sure they work well etc. And I tend to have a high standard for making sure I source things.
I personally think an LLM could help with some of this, and this is something I've been thinking about the past few days. But I'd have to build a pipeline and figure out a way to make it amplify what I like about my voice rather than have me speak through its voice.
I used to have a sort of puritanical view of art. And I think a younger version of myself would have been low key horrified at the amount of work in great art that was delegated to assistants. E.g. a sculptor (say Michelangelo) would typically make a miniature to get approval from patrons and the final sculpture would be scaled up. Hopefully for major works, the master was closely involved in the scaling up. But I would bet that for minor works (or maybe even the typical work) assistants did a lot of the final piece.
The same happens (and has always happened) with successful authors. Having assistants do bits here or there. Maybe some research, maybe some corrections, maybe some drafts. Possibly relying on them increasingly as you get later in your career or if you're commercially successful enough to need to produce at greater scale.
I think LLMs will obviously fit into these existing processes. They'll also be used to generate content that is never checked by a human before shipping. I think the right balance is yet to be seen, and there will always be people who insist on more deliberate and slower practices over mass production.
[0] Aside from internet comments of course, which are mostly stream of consciousness.
I like the perspective of "choices" during creation. It is an essential principle of the real art that it is a result of thousands/millions of deliberate choices. This is what we admire on the art. If you use mostly machine (or other kind of ways that decide instead and for you) for creation, you as an creator simply do less choices.
In this case, you delegate many of your experienced/crazy/hard decisions to the model (which is based on such decision made already by other artists but combines them in a random way). It is like decompressing JPG – some things are just hallucinated by machine.
From the perspective of pure human creativity, the result is thin, diluted. Even it seems like deliberate. In my opinion art lovers will seek for the dense art made by human, maybe asking even more for some kind of "proof" of the human-based process. What do you think?
What's interesting about thinking of code as art is that there rarely a variety of ways of implementing logic that's all optimal. So if you decide on the implementation and have a LLM code it, you likely won't need to make major changes given the right guidelines (I just mean like a single script, for the sake of comparison).
Writing is entirely different, and for some reason, generic writing even when polished (ChatGPT-esque tone) is so much more intolerable than say AI-generated imagery. Images can blend in the background, reading takes active processing so we're much more sensitive. And for the end user of a product, they care 0 or next to 0 about AI code.
> "Images can blend in the background, reading takes active processing so we're much more sensitive. And for the end user of a product, they care 0 or next to 0 about AI code."
AI is a tool like any other, and it can be used well or poorly, just like any other tool. It's important to know its limits. Being a tool, it must be studied for proper use.
Ai is useful in closed loop applications, often it can even do a decent job of closing the loop itself… but you need to understand that it is a fundamentally extractive, not creative, process. The body of human cultural knowledge is the underlying resource , and AI is the drill with which we pull out the parts we want.
Coding, robotics, navigation of constrained data spaces such as translation, tagging, indexing, logging, parsing, data transformations… those are all strong target candidates for transformer architecture automation.
As a professional writer, the author of this post is likely a better writer than 99.99% of the population. A quick skim of his blog suggests that he's comfortably more intelligent than 99% of people. I think it's totally unsurprising that he isn't fully satisfied with the output of LLMs; what is remarkable is that someone in that position still finds plenty of reasons to use them.
Now consider someone further down the scale - someone at the 75th, 50th or 25th percentile. The output of an LLM very quickly goes from "much worse than what I could produce" to "as good as anything I could produce" to "immeasurably better than anything I could hope to ever produce".
There have been quite a few skeptic blog posts recently about LLM. Some say they won't use it for coding, others for getting creative ideas, and others won't use it for editing and publishing. However, the silent issue all these posts have in common is that resistance is futile.
To be fair, I also don't like using Copilot when working on code. In many cases it turns into a weird experience when the agent generates the next line(s) and I basically become a discriminator judging if the thing really understands my problem and solution. To be honest, it's boring even if eventually it might make me turn in code faster.
With that said, I cannot ignore that LLMs are happening, and this is the future. The models keep improving but more importantly, the ecosystem keeps improving with things like MCP and better defined context for LLM tools.
We might be looking at a somewhat grim prospect. But like it or not, this is the future. Adapt and survive.
I understand. The question is what does it mean to "survive" for someone.
For me survival means:
- continuing to do my best at the language level – even if more people would start be gradually satisfied with less
- I just believe that education, critical thinking and evidence-based principles are at core of humanity progress and one day it will make comeback
- I am ok with smaller income and not wishing to exchange it for creating bullshit
The adaptation for me means:
- generally: stay open-minded
- I have to understand and somehow accept that the prospect is a bit grim but not to fall into some extreme and doom thinking
- I have to explore new ways how to augment human-oriented creativity (with or without these tools)
Same. I might use them for some things here and there, but not for writing. When I'm writing blog posts, people are coming to my articles to read what I've written, not what some glorified markov chain spits out.
> in a programming environment, you can immediately verify the answer by evaluating the code (at least for code snippets).
Well, it's a trap. You see a snippet is right, you accept it. Next time you do it faster, and faster. And then you get one that seems right but it's not. If you're lucky, it will cause an error.
24 comments
[ 2.7 ms ] story [ 46.3 ms ] threadI formed the habit of exporting entire chats to Markdown and found them useless. Whatever I found of useful from a given response either sparked a superseding thought of my own or was just a reiteration of my own intuitive thoughts.
I've moved from ChatGPT to Claude. The results are practically the same as far as I can tell (although my gut tells me I get better code from Claude) but the I think Anthropic have a better feel for response readability. Sometimes processing a ChatGPT response is like reading a white paper.
Other than that, LLMs get predictable to me after a while and I get why people suspect that they're starting to plateau.
That's AI.
I personally think an LLM could help with some of this, and this is something I've been thinking about the past few days. But I'd have to build a pipeline and figure out a way to make it amplify what I like about my voice rather than have me speak through its voice.
I used to have a sort of puritanical view of art. And I think a younger version of myself would have been low key horrified at the amount of work in great art that was delegated to assistants. E.g. a sculptor (say Michelangelo) would typically make a miniature to get approval from patrons and the final sculpture would be scaled up. Hopefully for major works, the master was closely involved in the scaling up. But I would bet that for minor works (or maybe even the typical work) assistants did a lot of the final piece.
The same happens (and has always happened) with successful authors. Having assistants do bits here or there. Maybe some research, maybe some corrections, maybe some drafts. Possibly relying on them increasingly as you get later in your career or if you're commercially successful enough to need to produce at greater scale.
I think LLMs will obviously fit into these existing processes. They'll also be used to generate content that is never checked by a human before shipping. I think the right balance is yet to be seen, and there will always be people who insist on more deliberate and slower practices over mass production.
[0] Aside from internet comments of course, which are mostly stream of consciousness.
I like the perspective of "choices" during creation. It is an essential principle of the real art that it is a result of thousands/millions of deliberate choices. This is what we admire on the art. If you use mostly machine (or other kind of ways that decide instead and for you) for creation, you as an creator simply do less choices.
In this case, you delegate many of your experienced/crazy/hard decisions to the model (which is based on such decision made already by other artists but combines them in a random way). It is like decompressing JPG – some things are just hallucinated by machine.
From the perspective of pure human creativity, the result is thin, diluted. Even it seems like deliberate. In my opinion art lovers will seek for the dense art made by human, maybe asking even more for some kind of "proof" of the human-based process. What do you think?
Writing is entirely different, and for some reason, generic writing even when polished (ChatGPT-esque tone) is so much more intolerable than say AI-generated imagery. Images can blend in the background, reading takes active processing so we're much more sensitive. And for the end user of a product, they care 0 or next to 0 about AI code.
Very interesting point!
Coding, robotics, navigation of constrained data spaces such as translation, tagging, indexing, logging, parsing, data transformations… those are all strong target candidates for transformer architecture automation.
Creative thought is not.
Now consider someone further down the scale - someone at the 75th, 50th or 25th percentile. The output of an LLM very quickly goes from "much worse than what I could produce" to "as good as anything I could produce" to "immeasurably better than anything I could hope to ever produce".
To be fair, I also don't like using Copilot when working on code. In many cases it turns into a weird experience when the agent generates the next line(s) and I basically become a discriminator judging if the thing really understands my problem and solution. To be honest, it's boring even if eventually it might make me turn in code faster.
With that said, I cannot ignore that LLMs are happening, and this is the future. The models keep improving but more importantly, the ecosystem keeps improving with things like MCP and better defined context for LLM tools.
We might be looking at a somewhat grim prospect. But like it or not, this is the future. Adapt and survive.
For me survival means: - continuing to do my best at the language level – even if more people would start be gradually satisfied with less - I just believe that education, critical thinking and evidence-based principles are at core of humanity progress and one day it will make comeback - I am ok with smaller income and not wishing to exchange it for creating bullshit
The adaptation for me means: - generally: stay open-minded - I have to understand and somehow accept that the prospect is a bit grim but not to fall into some extreme and doom thinking - I have to explore new ways how to augment human-oriented creativity (with or without these tools)
What do you think?
Well, it's a trap. You see a snippet is right, you accept it. Next time you do it faster, and faster. And then you get one that seems right but it's not. If you're lucky, it will cause an error.